CLEISTOGAMY 57 



Tradescantia erecta, Stellaria media, Spergula arvensis, Cerastium glomeratum, 

 Gaura parvifolia, Paronychia bonariensis, Corrigiola littoralis, Scleranthus annuus, 

 Herniaria glabra, Malva rotundifolia, and others ; according to Mechan (Bull. Torrey 

 Bot. CI., New York, x, 1883) in Nemophila maculata ^^/>^., Opuntia leptocaulis DC; 

 according to Coulter (Bot. Gaz., Chicago, viii, 1883) in Cyclamen europaeum; 

 according to Bush (op. cit., vii, 1882) in Malvastrum angustum; and according 

 to Battandier (Bui. soc. bot., Paris, xxx, 1883) in Portulaca oleracea Z., and others. 



These various forms of pseudo-cleistogamy point to the causes of genuine 

 cleistogamy : deficiency of light, air, warmth, dryness, or moisture are to be 

 regarded as such. Darwin referred the origin of cleistogamous flowers to 

 developmental retardation of chasmogamous forms, due partly to deficiency, but 

 also partly to an excess of light, and perhaps also to lack of insects (of. my note 

 on p. 54). 



Observation actually proves that even genuine cleistogamy may be produced 

 by deficiency of light. Kerner (' Nat. Hist. PI.,' Eng. Ed. i, II, p. 395), for instance, 

 noticed that Viola sepincola does not form any open flowers in the deep shades 

 of woods, but does form them in open country in situations that from time to 

 time receive sunlight. The investigations of Vochting (Jahrb. wiss. Bot., Leipzig, 

 xxv, 1893) agree with this, as he proved experimentally that with feeble illumination 

 there is frequent degeneration of the conspicuous parts, and formation of cleisto- 

 gamous flowers. And he also is of opinion that defective illumination is of primary 

 importance in the evolution of cleistogamy. 



The question whether there are plants which bear cleistogamous flowers only was 

 at first answered in the affirmative. A. Batalin (Bot. Ztg., xxix, 1871, pp. 388-92) 

 maintained that Juncus bufonius is regularly self-fertilized, its flowers always remaining 

 closed. It was, however, established by Ascherson (op. cit., 187 1, pp. 551-5; 1872, 

 pp. 697-9, 738) 739), Buchenau (op. cit, 1871, pp. 845-52), and Haussknecht 

 (op. cit., pp. 802-7) that while numerous flowers fade with enclosed pistil and 

 anthers, others on the contrary open out to 180 in a stellate way, so that they 

 may also be fertilized by foreign pollen ; and it was also shown that between 

 these two kinds of flower there are numerous intermediate forms (cf. F. Buchenau, 

 * Bestaubungsverhaltnisse der Juncaceen,' Jahrb. wiss. Bot., Leipzig, xxiv, 1892, 

 PP- 363> 364, 382). 



There is also an African species of Salvia which was at one time supposed 

 to bear cleistogamous flowers only. It was therefore named S. cleistogama de Bary 

 et Paul, and was regarded by Ascherson (Bot. Ztg., xxix, 187 1) as an example of 

 a plant propagating continuously by self-fertilization. As a matter of fact, the plant 

 bore cleistogamous flowers only during the first five years of its cultivation at Halle, 

 but afterwards chasmogamous flowers appeared as well. 



Recently (1883) some cases of the constant occurrence of completely closed, 

 but otherwise normally formed flowers have been announced by W. Burck (Ann. 

 Jard. bot., Buitenzorg, iv, pp. 17-20). In Myrmecodia echinata Gaud., e.g., the 

 flowers are completely closed by fusion of the four corolla-lobes, although there are 

 secreting nectaries developed beneath a ring of hairs. The stigmas, which are 

 papillose on the outer side, alternate with the anthers, and by growth of the 

 corolla the pollen of these anthers is brought into contact with the stigmatic 



